论瓦拉赫与艾伦的AMAs的伦理设计思想——兼评《机器伦理:教导机器人区分善恶》

Wallach and Allen's Ethics Thought of AMAs Design——Review of Moral Machines:Teaching Robots Right from Wrong

随着人工智能技术的发展,(技术)人工物道德行为体(AMAs)日益呈现出高度的"自主权"和"感受性",使得机器(人)走出科幻小说的窠臼。由于世界需要善建AMAs、必要的道德"图灵测试"、强人工智能成为必然趋势,这些成为善建AMAs的时代性要求。瓦拉赫与艾伦提出了"三重区分"理论对AMAs进行了伦理定位,并提出了"自上而下式(TD)"和"自下而上式(BU)"两条伦理设计方法。瓦拉赫与艾伦立足于"未来场景"所进行的AMAs伦理设计思想为未来机器(人)的发展奠定了道德性根基。

Along with the development of artificial intelligence technology , artificial moral agent （AMAs） in-creasingly shows a high degree of autonomy and sensibility , making machine （robot ） jump out of science fiction and come true .The reasons are that the world needs good AMAs , necessary moral “Turing test” and the strong ar-tificial intelligence becomes an inevitable trend which also becomes a requirement for the era .Wallach and Allen put forward a “triple distinction” theory on the AMAs ethics orientation , and two methods of ethical design , i.e.“top-down type （ TD）” and “from bottom to top type （ BU） .” Based on the “future scenarios ,” Wallach and Al-len conducted ethical design thought of AMAs which establishes ethical foundation for the development of the future machine （robot）.